ProLiant DL380 G5 unexpected Shutdown

[jones1313]

Hi,

we have a ProLiant DL380 G5 that shows a corrupted SQL DB after unexpected Reboot.

Two things that i could determine:

-Post Error - Message: 1794-Drive Array - Array Accelerator Battery Charge Low - Error: 94

-System Crash - Message: BugCheck, STOP: 0x0000007A (0xC03DC71C, 0xC000000E, 0xF71C7424, 0x00F7E860)

-640004: Controller has reported a critical performance threshold error on this drive

Is it likely, that the bad drive will cause that blue screens, and therefore the unxpected shutdowns?

Furthermore: How to replace the backup battery e.g. accelerator board?

[Murali_V]

Hi, 

POST error 1794 is for Drive Array:

  Array Accelerator Battery Charge Low Array Accelerator is temporarily disabled.

  Array Accelerator will be re-enabled when battery reaches full charge.

  Possible Cause: The battery charge is below 75 percent. Posted writes are disabled.

  Solution: Replace the array accelerator board if the batteries do not recharge within 36 powered-on hours.

Also,

Make sure that the Controller firmware and drivers are up-to-date.

Run Array diagnostics to check for any failed drive as you are suspecting.

Regards, 

Murali V

[Murali_V]

HP ProLiant DL380 Generation 5 Server Maintenance and Service Guide Page number 53 for Battery-backed write cache procedures:

http://h20566.www2.hp.com/hpsc/doc/public/display?sp4ts.oid=1121516&docId=emr_na-c00710359&docLocale=en_US

[jones1313]

Thanks for that!

After a reboot, the battery issue disappeared.

HP Insight Diagnostics give a warning regarding the hard-drive, but the ADU Report looks good.

Im still troubleshooting for the unexpected reboot, that occured a few times now. I cant determine a windows process for that. Also the UPS was working steady.

Do you have one more good hint regarding the unexpected reboots?

Many thanks so far!

[Murali_V · Accepted solution]

Hi, 

Check Integrated Management Log (IML) and Windows Event Logs for the reason for restart. 

Check the logs at the time of restart. 

[jones1313]

Have done this! The restart is absolutely out of the blue, without any prior events in the log! I expect that is is from a hardware problem, but im not that experienced in HP Servers yet.

[Torsten.]

Did you check the ILO log too?

[PZel]

The Smart controller is it a E200  or a P400/800  ??

or does the server has another RAID controller in it ??

The BSOD (0x7A  (0C000E)) is referencing to a hardware SCSI problem (bad disk or bad controller).

Whenever its a E200 then there might be a problem with the BBWC piggy back card (SPN 413486-001).

You may try to reseat or swap this little card.

Whenever its  a P400 then there might be also problems with a small piggyback board mounted on the P400.

You can reseat or change that.

Normally the battery warning is because the server is suddenly going down and then up again, so its probably  not the cause of the problem.

[jones1313]

Hi PZel,

it is a P400, no other Raid Controller installed here!

Should i replace the whole P400, or is it possible to get only the Piggyback board?

Many thanks for the explanation regarding the Battery warning!

[PZel]

We use here a lot the 256MB BBWC for P400: SparePart: 405836-001, but is also possible that your BBWC is a 512MB version: 405835-001. You can peel it off from the P400 controller (use wristband).

Reseating the BBWC is sometimes the solution.

It is also possible that the P400 itself is bad. (SPN 405831-001).

Do you have another DL380G5 in the neighbourhoud for (temporarely) exchanging parts ??